Discussion
The management of locally advanced laryngeal squamous cell carcinoma (LA-LSCC) has evolved to favor combined treatment methods for better patient outcomes, often integrating surgery, radiotherapy, chemotherapy, targeted therapies, and immunotherapy [
14]. Concurrent chemoradiation is now standard for advanced-stage disease, aiming to reduce tumor size and eliminate micrometastatic disease. This approach has significantly improved laryngeal cancer management by enabling laryngeal preservation and enhancing quality of life [
15].
Our findings demonstrate the effectiveness of combining chemotherapy and radiotherapy in reducing tumor burden and achieving high laryngeal preservation rates in patients with LSCC, highlighting the synergistic effects of this treatment strategy. By preserving the larynx in most patients, this approach improves oncological outcomes and helps maintain essential functions such as speech and swallowing, enhancing overall quality of life. These results support the use of induction chemotherapy followed by radiotherapy as a valuable treatment option for LSCC and are consistent with findings from other studies. Specifically, our data shows that while induction chemotherapy (IC) alone achieved a 31% complete response (CR) and 49% partial response (PR) in 81 patients, the subsequent addition of radiation therapy (RT) in 65 patients dramatically increased the CR rate to 74% and maintained a 20% PR rate. This highlights the synergistic effects of this sequential treatment strategy. By preserving the larynx in most patients, this approach improves oncological outcomes and helps maintain essential functions such as speech and swallowing, enhancing the overall quality of life. These results support the use of induction chemotherapy followed by radiotherapy as a valuable treatment option for LA-LSCC and are consistent with findings from other studies [
16,
17,
18,
19].
In this context, the identification of novel, clinically relevant biomarkers for early diagnosis, accurate prognosis, and effective treatment response prediction remains a critical need for optimizing patient care in locally advanced laryngeal squamous cell carcinoma (LA-LSCC). The present study was specifically designed to investigate the potential of microRNA-449a (miR-449a) as such a biomarker. Our objectives were to evaluate its diagnostic utility by analyzing expression profiles to discriminate between patients with LA-LSCC and healthy control subjects, and to explore its prognostic significance by examining correlations with pertinent clinicopathological features and patient outcomes. Through a comprehensive evaluation of miR-449a expression patterns, this research aimed to validate its suitability as a non-invasive biomarker for LA-LSCC diagnosis and to assess its capacity to yield valuable prognostic information that could contribute to refined risk stratification and personalized treatment strategies.
Our findings establish miR-449a as a potential diagnostic biomarker for LA-LSCC. We observed a significant downregulation of miR-449a expression in LA-LSCC tissues compared to adjacent normal laryngeal tissues (p < 0.0001). This pattern aligns with the widely recognized role of miR-449a as a tumor suppressor in various cancers, where its diminished expression can promote uncontrolled cell proliferation and tumor progression [
12,
20,
21]. Extending these findings to a non-invasive context, we also detected a statistically significant downregulation of miR-449a in the serum of LA-LSCC patients compared to healthy controls (p < 0.0001), highlighting its potential as a circulating biomarker. Regarding diagnostic accuracy, ROC curve analysis demonstrated that tissue miR-449a possessed good discriminatory power, with an Area Under the Curve (AUC) of 0.857 (p < 0.0001) for distinguishing malignant LSCC tissues from normal tissues. While serum miR-449a showed a moderate diagnostic accuracy (AUC = 0.734, p < 0.0001) for detecting LA-LSCC in patient serum, its non-invasive nature makes it a highly attractive candidate for initial screening or disease monitoring. Importantly, a strong positive correlation (R² = 0.988, p < 0.0001) was observed between miR-449a levels in LA-LSCC tissues and their corresponding serum samples. This robust correlation suggests that serum miR-449a can reliably reflect the expression status within the tumor, further supporting its utility as a non-invasive surrogate biomarker for LA-LSCC. These results are consistent with a growing body of literature on microRNAs as promising non-invasive diagnostic tools in head and neck cancers. To the best of our knowledge, this is the first study to validate miR-449a’s diagnostic and prognostic role specifically in LSCC, particularly in locally advanced disease.
Our analysis explored the association of miR-449a expression with various clinicopathological characteristics. Univariate analysis indicated that higher miR-449a expression was significantly associated with younger age, female sex, non-smoking status, and less aggressive disease features (lower tumor grade, earlier T-stage, limited nodal involvement, and earlier TNM stage). However, the multivariate multiple regression analysis, which provides a more robust assessment of independent predictors by controlling for other variables, revealed nuanced insights into factors influencing miR-449a expression. In this model, female sex (p = 0.014), older age (p = 0.034), more advanced T-stage (p = 0.049), more advanced N-stage (p = 0.001), and higher TNM stage (p = 0.002) were all independently associated with higher miR-449a expression. Conversely, smoking status was a significant negative predictor (p < 0.0001), meaning smokers had lower miR-449a levels. Pathological grade and performance status did not show significant independent associations. The difference between univariate and multivariate findings underscores the complexity of biological systems and the importance of accounting for confounding factors. While the univariate analysis might suggest an inverse correlation with disease aggressiveness, the multivariate analysis implies that in the context of other variables, miR-449a expression can increase with certain aspects of disease progression, potentially reflecting a compensatory mechanism or involvement in complex cellular responses to advanced disease.
Beyond its diagnostic potential, our study firmly establishes miR-449a as a significant prognostic factor in LA-LSCC. Kaplan-Meier survival curves clearly demonstrated that patients with high miR-449a expression exhibited significantly prolonged overall survival (median 67.82 months vs. 23.74 months; Log-rank p = 0.0012) compared to those with low expression. A similar trend was observed for disease-free survival, where high miR-449a expression was associated with significantly better outcomes (median 66.03 months vs. 23.74 months; Log-rank p = 0.0002). The multivariate Cox proportional hazards regression analysis (
Table 8) further strengthened these findings by identifying high miR-449a expression as an independent predictor of improved overall survival (Hazard Ratio [HR] = 0.15, 95% CI: 0.055–0.410, p < 0.001). This indicates a strong protective effect of elevated miR-449a on survival outcomes in LA-LSCC. These results are consistent with previous research indicating that miR-449a can be a valuable tool to predict cancer-specific survival outcomes in patients with different types of cancer [
12,
22,
23]. Consistent with known prognostic factors in cancer, our analysis also confirmed that older age (p = 0.004), smoking status (p = 0.017), higher tumor grade (p = 0.003), and more advanced T-stage (p = 0.001), N-stage (0.006), and TNM stage (p < 0.001) were all independently associated with an increased hazard of mortality. The robust independent prognostic value of miR-449a suggests its utility in refining risk stratification for LA-LSCC patients, potentially guiding more personalized therapeutic strategies and follow-up schedules.
Our investigation into miR-449a expression dynamics during treatment provides valuable insights into its potential as a predictor of therapeutic response. We observed a consistent and statistically significant increase in miR-449a expression levels after both induction chemotherapy and subsequent radiation therapy in patients who achieved either a complete or partial response (all p < 0.0001). For instance, in patients achieving complete response to induction chemotherapy, miR-449a significantly increased from 2.369 ± 0.416 to 3.948 ± 0.491. Similarly, complete responders to radiation therapy showed an increase from 2.809 ± 0.421 to 3.472 ± 0.401. This suggests that miR-449a levels dynamically change in response to effective treatment, potentially reflecting tumor cell apoptosis, reduced tumor burden, or other favorable biological responses. This finding is particularly exciting as it positions miR-449a as a potential real-time biomarker for monitoring therapeutic efficacy, which could inform adaptive treatment protocols and improve patient management. Such dynamic changes in miR-449a expression in response to treatment have also been observed in other cancers; for example, a study in osteosarcoma patients found that miR-449a levels were significantly higher after surgery in responders, suggesting its role in assessing treatment success [
24].
Despite the compelling findings, our study has certain limitations. The cohort size of 81 patients, while sufficient for statistical significance in many analyses, may limit the generalizability to broader LA-LSCC populations. As a single-center study, our findings warrant validation in larger, multi-institutional, prospective cohorts to confirm their broader applicability and clinical utility. Further research should also focus on elucidating the precise molecular mechanisms by which miR-449a exerts its observed effects in LA-LSCC. This includes identifying its direct mRNA targets and the downstream signaling pathways involved in cell cycle regulation, apoptosis, and epithelial-mesenchymal transition, which could explain both its tumor-suppressive role and its dynamic changes with treatment. Functional studies in vitro and in vivo, such as overexpression or knockdown experiments of miR-449a in LSCC cell lines and animal models, are crucial to establish a causal relationship between miR-449a levels and tumor behavior or treatment response. Ultimately, combining miR-449a with other promising biomarkers or integrating it into multiparametric panels could further enhance its diagnostic and prognostic accuracy in LA-LSCC.